The chairs have spoken - no change on extensions stance in HTML

I'll try to track down the email from the HTML-XML task force, but the HTML
WG chairs have handed down the decision that because no use case provided a
sufficiently compelling exception, the decision of the HTML 5 working group
is that no extension mechanisms will be considered, and explicitly no
namespaced extension mechanisms will be considered as part of HTML. This
means that if XForms is in an HTML page, then such a page will not validate
as HTML. No real surprise there. What was unclear was the role of processing
instructions, and most specifically the role of the <?xml-stylesheet?> as
part of this process, though the minimalist approach that the HTML working
group is taking raises questions about the strength of support for that.
This is probably something that needs ot be surfaced on the HTML/XML task
force.

Here's my take of this on the implications for XForms:

   - XForms in HTML5 will be considered non-validating in an HTML parser.
   Any use of namespace designators - such as xmlns:a or <a:foo> will be
   treated as if these were simply attribute or element names with no specific
   binding (that is to say <a:foo><a:bar>This is bar</a:bar>This is foo</a:foo>
   will not be treated as being in namespace, but rather as elements that
   happen to have the same starting characters "a:" This falls into the
   "oh-well" category - this was pretty much expected, and can be worked
   around.
   - XForms as it sits right now is strictly speaking an XML construct -
   that is to say, it is possible that a mixed schema can be prepared for
   XFORMs, but this schema will be the relevant one on acting on XHTM. One
   consequence of this is that it probably raises the prominence of
   establishing an equivalency relationship between the various schema
   proposals and establishing XHTML as the default host language, before XHTML
   gets locked down in any way.
   - I'm not sure about where things stand with regard to WebBL vs XBL -
   sounds like it should be a topic of discussion for tomorrow based upon the
   meeting last Friday.

In short, not much has changed, other than the near certainty that XForms on
HTML will be a non-starter in its present form.

Kurt

Kurt Cagle
kurt.cagle@gmail.com
443-837-8725